public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com>
To: Tahsin Erdogan <tahsin@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
	tglx@linutronix.de, mingo@redhat.com, hpa@zytor.com,
	x86@kernel.org, Waiman.Long@hp.com, borntraeger@de.ibm.com,
	oleg@redhat.com, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] x86/spinlocks: Fix regression in spinlock contention detection
Date: Tue, 05 May 2015 16:08:22 +0530	[thread overview]
Message-ID: <55489D9E.4010003@linux.vnet.ibm.com> (raw)
In-Reply-To: <20150505091714.GF21418@twins.programming.kicks-ass.net>

On 05/05/2015 02:47 PM, Peter Zijlstra wrote:
> On Mon, May 04, 2015 at 09:15:31PM -0700, Tahsin Erdogan wrote:
>> A spinlock is regarded as contended when there is at least one waiter.
>> Currently, the code that checks whether there are any waiters rely on
>> tail value being greater than head. However, this is not true if tail
>> reaches the max value and wraps back to zero, so arch_spin_is_contended()
>> incorrectly returns 0 (not contended) when tail is smaller than head.
>>
>> The original code (before regression) handled this case by casting the
>> (tail - head) to an unsigned value. This change simply restores that
>> behavior.
>>
>> Fixes: d6abfdb20223 ("x86/spinlocks/paravirt: Fix memory corruption on
>> unlock")
>> Signed-off-by: Tahsin Erdogan <tahsin@google.com>
>> ---
>>   arch/x86/include/asm/spinlock.h | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/arch/x86/include/asm/spinlock.h b/arch/x86/include/asm/spinlock.h
>> index cf87de3..64b6117 100644
>> --- a/arch/x86/include/asm/spinlock.h
>> +++ b/arch/x86/include/asm/spinlock.h
>> @@ -169,7 +169,7 @@ static inline int arch_spin_is_contended(arch_spinlock_t *lock)
>>   	struct __raw_tickets tmp = READ_ONCE(lock->tickets);
>>
>>   	tmp.head &= ~TICKET_SLOWPATH_FLAG;
>> -	return (tmp.tail - tmp.head) > TICKET_LOCK_INC;
>> +	return (__ticket_t)(tmp.tail - tmp.head) > TICKET_LOCK_INC;
>
> I'm not seeing it, everything in that expression is of __ticket_t type
> (tail, head and TICKET_LOCK_INC), nothing should cause it to be cast to
> another type due to conversion rules.
>
> Or does - always cast to a signed type? Lemme go grab the C rules again.
>
> I'm not seeing it.. Please explain better, iow. your changelog fails to
> properly explain the problem.
>

Same from me here.

Did you really see a regression?

I am not seeing two unsigned value subtraction resulting in a negative
value.

================
#include <stdio.h>

#define LOCK_INC 2

int main()
{
         unsigned int head = 32700, tail=2;

         if ((tail - head) > LOCK_INC)
                 printf(" tail - head > LOCK_INC \n");
         else
                 printf(" tail - head < LOCK_INC \n");

         return 0;
}



  reply	other threads:[~2015-05-05 10:34 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-05-05  4:15 [PATCH] x86/spinlocks: Fix regression in spinlock contention detection Tahsin Erdogan
2015-05-05  9:06 ` [tip:x86/urgent] " tip-bot for Tahsin Erdogan
2015-05-05  9:17 ` [PATCH] " Peter Zijlstra
2015-05-05 10:38   ` Raghavendra K T [this message]
2015-05-05 10:58     ` Peter Zijlstra
2015-05-05 14:10       ` Tahsin Erdogan
2015-05-05 14:30         ` Peter Zijlstra
     [not found]       ` <CAAeU0aPSDdDVKm=YgbYH+SWfk07rLGgEFMQoN+nt4vc1_YLHzg@mail.gmail.com>
2015-05-05 15:25         ` Raghavendra K T
2015-05-05 15:28           ` Tahsin Erdogan
2015-05-05 15:32           ` Waiman Long
2015-05-05 15:38             ` Raghavendra K T
2015-05-06 15:37 ` Raghavendra K T
2015-05-07  7:13   ` Tahsin Erdogan
2015-05-08 11:09     ` Greg KH

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=55489D9E.4010003@linux.vnet.ibm.com \
    --to=raghavendra.kt@linux.vnet.ibm.com \
    --cc=Waiman.Long@hp.com \
    --cc=borntraeger@de.ibm.com \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=oleg@redhat.com \
    --cc=peterz@infradead.org \
    --cc=tahsin@google.com \
    --cc=tglx@linutronix.de \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox